The process of somatic hypermutation (SHM) of immunoglobulin (Ig) genes is initiated by the cytidine deaminase AID creating cytidine (C) to uridine (U) transition mutations. Outside of SHM, U is faithfully repaired to C with the help of base excision repair and DNA mismatch repair. During SHM, paradoxically, these repair mechanisms are recruited together with translesion DNA polymerases to create mutations at and near the U by error-prone repair. This grant application proposes to study two major questions concerning the SHM process. One, how are the mutations restricted to about 2 kb from the promoter and sparing the transcribed 5< 100 bp of the Ig genes? We have proposed a model where AID loads on the transcription complex near the promoter, travels with the RNA polymerase, and stochastically dissociates from the complex while deaminating cytosines in its track. Once dissociated, AID cannot re-load onto the transcription machinery. An alternative model is that AID is attracted loosely to Ig loci (due to cis-elements that we have recently identified as essential, in combination with elements in the Ig enhancers) but does not travel with transcription complexes. We plan to determine the location of AID and RNA polymerase on the AID target gene during SHM. We will also follow up on preliminary findings that AID indeed requires the transcription process itself. We will further investigate the role of nucleosomes that appear to be a barrier to SHM. These experiments will be important to learn how SHM can create variable region mutations whereas the constant region of Ig genes is spared from mutations, facts that are essential to enhance and protect the defensive and physiological roles of the variable and constant regions, respectively. The other question relates to new findings by others and our lab that a considerable proportion of uracils are created by AID in non-Ig genes as well as in Ig genes and that many of the AID-induced uracils can be repaired error-free in both types of genes, while another proportion is treated in an error-prone fashion. We have shown, that error-free repair of AID lesions does occur in Ig genes as well as in non-Ig genes. We propose to produce mutating B cell lines that express modified AID molecules that are either expressed only during the G1- phase or only during S, G2, and M. The time in the cell cycle when AID is expressed will affect which other proteins interact with AID and with the AID-induced uracil, thus determining whether the uracil and neighboring sequences become mutated or are repaired error-free. Most of the experiments will at first be carried out in B cell lines in culture that undergo somatic hypermutation. If conclusive in cultured cells, some experiments will be checked in normal and transgenic mice.